Could DNA have kick-started life on Earth instead of RNA?

By Philip Ball

Researchers have managed to create bits of DNA that can stitch themselves together without a helping hand from other molecules. By contrast, natural DNA needs enzymes to stitch itself up, correct mutations, or make copies of itself.

The creation of this super-capable DNA suggests that rare bits of natural DNA might have evolved the same capability in the past. That could alter our thinking about how life began.

Researchers have known for decades that RNA, the molecule that translates information in DNA into proteins, can show enzyme-like behaviour. These special forms of RNA, called ribozymes, won their discoverers a Nobel Prize in the 1980s. Since then, several ribozymes have been found in cells.

Ronald Breaker of Yale University and his colleagues have shown that the equivalent molecules for DNA, deoxyribozymes, can be created. "We have shown that RNA is not special in that regard," says Breaker.

Now they have gone one step further, devising deoxyribozymes that catalyze links between DNA molecules themselves, as they report in the Journal of the American Chemical Society1. Creation of these links is an essential part of the process by which DNA replicates itself.

If natural DNA could do that too, it might have been able to kick-start life on this planet.

Many researchers reckon that RNA, rather than DNA, formed the first building block of life, in part because it has this capacity to behave like an enzyme. There are many other pieces of evidence to indicate that life on our planet began with RNA, but it now seems that this molecule's priority may owe more to chance than to any special capabilities.

"If you play out the origin of life on a thousand different planets like Earth it may sometimes look very different," says Breaker, suggesting our world might begin with DNA rather than RNA.

Slow going

Breaker's team made the deoxyribozymes by artificially evolving natural DNA in a test-tube. Strands of DNA were shuffled at random. Those that showed enzyme-like activity were picked out and fine-tuned by repeating the shuffling process over and over. The researchers found they needed two kinds of deoxyribozyme to stitch DNA together.

So far the deoxyribozymes work 100,000 times slower than the natural enzyme, called DNA polymerase. But if this can be improved, then deoxyribozymes could benefit biotechnologists.

Jack Szostak of the Massachusetts General Hospital in Boston thinks this might be possible. "I would not be surprised if DNA can do anything RNA can do," he says.

Researchers are exploring the use of artificially evolved RNA molecules to make drugs and to detect other molecules such as toxins. RNA is useful because it does not need an enzyme to glue itself together. But there are some disadvantages: it has a tendency to fall apart in water, and in some ways it is harder to make than DNA.

Researchers would prefer to use DNA if they could, says Szostak, and deoxyribozymes might give them that chance.

References 1. Sreedhara, A., Li, Y. & Breaker, R. R. . Journal of the American Chemical Society, 126, 3454 - 3460, doi:10.1021/ja039713i (2004). |Article|

Contact: Lauren Ward wardle@andrew.cmu.edu 412-268-7761 Carnegie Mellon University